DESCRIPTION (Applicant's Description): He patocellular carcinoma (HCC) is one of the most common and fatal malignancies worldwide. In the last decade, there has been remarkably little progress in the development of novel therapeutic strategies for this disease. Certain features of this cancer (alpha-fetoprotein production, p53 status) may make it amenable to gene therapy approaches. The overall goal of this project is to develop adenovirus based strategies for the treatment of hepatocellular carcinoma. The systemic administration of cytokines in cancer therapy is limited by toxicity to the host. An alternative strategy is direct tumor transduction with adenoviral vectors engineered to produce immunoregulatory cytokines. H o w ever, intratumor administration of adenovirus results in systemic distribution and expression of the vectors; therefore, for proinflammatory cytokines, transcriptional targeting is essential. The alpha-fetoprotein gene regulatory elements will be utilized to construct transcriptionally specific adenoviral vectors engineered to deliver cytokine genes, and these vectors will be tested in an animal model of HCC. In the second adenoviral-based strategy, mutations of the p53 tumor suppressor gene, which are commonly present in hepatocellular carcinomas, will be utilized. D11520 is an adenovirus that has a mutation which limits its ability to replicate to p53(-) cells. The ability of d11520 to selectively replicate and lyse p53(-) HCC cells will be studied and applied to the treatment of HCC.